Flex and Priority Inference Tiers Introduced

Introduced the new Flex and Priority inference tiers, offering more options for optimizing cost or latency.